Session 10.3 INM RAS coupled Arctic Ocean\ sea ice model. The results of the AOMIP 31-year coordinated spin-up 1948–1978

Thursday, 15 May 2003: 10:15 AM
Nikolai G. Yakovlev, Institute of Numerical Mathematics, Russian Academy of Sciences, Moscow, Russia

Presentation PDF (585.3 kB)

The development of the mathematical theory of climate and corresponding numerical methods of climate modeling is one of the topics for the Institute of Numerical Mathematics of Russian Academy of Sciences (INM RAS). On the basis of the extended experience in 3D large-scale ocean climate modeling the new coupled ocean\sea ice model has been developed for Arctic Ocean and polar seas climate studies.

The physical background of the model is the hydrostatic primitive equation ocean model with the linearized kinematic condition at the upper surface, Parkinson and Washington style model for ice\snow thermodynamics and viscous-plastic rheology for ice dynamics. There're several gradations of the ice thickness, with the simple parameterization of the ice thickness redistribution during ridging. For the river run-off both mass and fresh water fluxes are taken into account. Model domain covers area of Arctic Ocean and GIN Sea north 65N. This version of the model is aimed mostly to methodological experiments, so the spatial grid size is approx. 100 km.

The numerical scheme of the model is based on the finite-element (FE) spatial approximation. Time approximation is made by the time-splitting scheme with some special treatment of nonlinear sea ice rheology. There the step of vertical turbulent diffusion of temperature and salinity is extracted, when snow\ice thermal evolution and vertical profiles of temperature and salinity over the whole depth are determined simultaneously by the implicit time scheme. Wind drift problem is solved for water and ice current velocities in a similar way. Temperature, salinity and momentum advection is approximated by the FE upwind scheme with no crosswind diffusion by Hughes and Brooks.

This model was modified to adopt the Arctic Ocean Model Intercomparison Project (AOMIP) forcing and parameterizations. Model is driven by realistic NCEP\NCAR Reanalysis daily air temperature and sea level pressure, as well as climatological monthly means of river run-off, precipitation, cloudiness, humidity and ocean temperature and salinity at open boundaries. The first stage of the AOMIP is the 30-year Coordinated Spin-up 1948-1977.

The results on the 30-year spin-up of the Arctic Ocean climate system are presented and compared with observations. The spatial distribution and temporal variability of ocean temperature and salinity, sea level and ice characteristics are investigated with the special focus on the Atlantic water pathways and Arctic Ocean freshwater content. The limits and utility of the coarse resolution models in Arctic modeling are also discussed.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner